|Publication number||US7715531 B1|
|Application number||US 11/173,804|
|Publication date||May 11, 2010|
|Filing date||Jun 30, 2005|
|Priority date||Jun 30, 2005|
|Publication number||11173804, 173804, US 7715531 B1, US 7715531B1, US-B1-7715531, US7715531 B1, US7715531B1|
|Inventors||Andrew R. Golding, Egon Pasztor|
|Original Assignee||Google Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (20), Non-Patent Citations (1), Referenced by (27), Classifications (15), Legal Events (2)|
|External Links: USPTO, USPTO Assignment, Espacenet|
Interactive voice response systems, often referred to as voicemail systems, are a common way of channeling incoming calls. Many voicemail systems can be inefficient and annoying for the caller who may need to listen to multiple levels of options before reaching his final destination. It would be beneficial to have a more efficient way of navigating through a voicemail system.
Various embodiments of the invention are disclosed in the following detailed description and the accompanying drawings.
The invention can be implemented in numerous ways, including as a process, an apparatus, a system, a composition of matter, a computer readable medium such as a computer readable storage medium or a computer network wherein program instructions are sent over optical or electronic communication links. In this specification, these implementations, or any other form that the invention may take, may be referred to as techniques. A component such as a processor or a memory described as being configured to perform a task includes both a general component that is temporarily configured to perform the task at a given time or a specific component that is manufactured to perform the task. In general, the order of the steps of disclosed processes may be altered within the scope of the invention.
A detailed description of one or more embodiments of the invention is provided below along with accompanying figures that illustrate the principles of the invention. The invention is described in connection with such embodiments, but the invention is not limited to any embodiment. The scope of the invention is limited only by the claims and the invention encompasses numerous alternatives, modifications and equivalents. Numerous specific details are set forth in the following description in order to provide a thorough understanding of the invention. These details are provided for the purpose of example and the invention may be practiced according to the claims without some or all of these specific details. For the purpose of clarity, technical material that is known in the technical fields related to the invention has not been described in detail so that the invention is not unnecessarily obscured.
If it is determined not to call this number (202), then N is set to equal to N+1 (204), and it is determined whether the next number should be called (202).
If this phone number is called, then it is determined whether it is answered by a voicemail system (206). If it is not answered by a voicemail system, then this phone number is added to a list of numbers not to spider (212). For example, if the phone call was answered by a person or an answering machine then that phone number may be added to a list of phone numbers not to chart. Alternatively, the phone number may be deleted from a list of phone numbers to call.
If a voicemail system is reached (206), then the voicemail choices offered by the voicemail system is charted (208). Further details of charting the voicemail choices are later discussed in conjunction with the remaining figures. The database of phone numbers is updated with the charted information (210). For example, information such as confirmation of the existence of the business may be associated with the phone number, or a graphical representation of the voicemail choices may be associated with the business that owns the phone number.
It is then determined whether to call this number (302). If it is determined not to call this number, a new phone number is selected (300). An example of why a phone number would not be called is if that phone number is listed on a “do not call” list. An example of how phone numbers are added to such a list is that the owner of a phone number may have voluntarily put the number on this list. Another example is that phone numbers may be added to this list if it is determined that these numbers are currently inappropriate for automatic voicemail exploration.
Once a phone number is called (302), then it is determined whether a person answers the call (304). If a person answers the call, then procedures for a person answering the call can be followed (306). For example, if a person answers the call, the call may be terminated. In another example, a conversation may ensue, such as automatically asking the person to verify that this number is for the name of the business listed for this number. Thereafter, information regarding this phone number is updated (308). For example, it may be noted that the business listed for this number has been confirmed to still be true. Another piece of information to be updated may be that there is no voicemail system for this business and that this phone number should be added to the “do not call” list. Thereafter, a new phone number is selected (300).
If a person does not answer the call (304), then it is determined whether a machine answers the call (310). An example of how a determination of whether a person answers a call, or an answering machine answers the call or a voicemail system answers the call may be performed by modeling a natural human answering a phone versus a machine voice answering the phone versus an answering machine message. For example, a Hidden Markov Model (HMM) may be used to build such models. Hidden Markov Models are well understood by persons in the art. For example, a person answering the phone may be distinguished from a voicemail system by noted factors such a rise in the inflection, or mumbling, or recognizing certain phrases, such as “Hello?” versus “Hello, you have reached . . . ”.
If a person does not answer the call, and a machine also does not answer the call (310), then the call is terminated and it is noted that there is no answer at this number (316). This information is updated and associated with this phone number (318), and a new phone number is selected for analysis (300).
If a machine answers the call (310), then it is determined whether it is answered by an answering machine (312). An answering machine as used herein includes any automatic phone answering system that does not offer a selection of choices. Examples of answering machines include a personal answering machine or an automatic message. Examples of an automatic message include a phone forwarding message, a message stating that the number has changed and providing a new number, or a message stating that this number no longer exists.
If the call is answered by an answering machine (312), then procedures for the call being answered by an answering machine are followed (314). An example of such a procedure is recording the message spoken on the answering machine for further analysis. For example, a human may listen to the recorded message to determine further action regarding this phone number. In another example, certain phrases or words may be recognized such as “forwarding”, “new number”, or “no longer exists”, or “your call cannot get through”. Appropriate information, such as “forwarded”, “new number”, “no longer exists”, may be updated regarding this phone number (308). Thereafter, a new phone number is selected for analysis (300).
If a machine answers the call (310), but it is not an answering machine (312), then it is determined whether a choice is given by the recorded response (330). For example, a voicemail system may start with a choice of pressing 1 to continue in English. If no choice is given (330), then procedures for an answering machine are followed (314). If, however, at least one choice is given, then one of the options is selected for analysis (332). For example, the number 1 option may be analyzed.
It is then determined whether this option goes back to a previous menu (334). In one example, this determination can be made by actually selecting the option and recognizing that the recording has returned to a previous menu. In another example, the analysis may occur prior to actually choosing the option and recognizing certain words such as “to go back” or “previous menu”. If this option does go back to a previous menu (334), then this option is skipped (336), and a new option is selected for analysis (332).
In one embodiment, information is updated with regard to the phone number at every level within the voicemail system. For example, whenever at least one choice is offered, the choice is saved, for example, by recording the choice. This recorded choice may be literally recorded so that the choice can be heard by a human, or in another example, the choice can be saved in a database with information associated with the offered choice. For example, if the choice that it is given is “for English, press 1” then the phrase “for English” and “press 1” may be recognized and the information saved on the database may be “English=1”. If the choice is audibly recorded, then it may be transcribed into text for further use, by for example, manual transcription or automatic speech recognition. In one example, automatic speech recognition programs such as IBM's Via Voice or Dragon maybe used in conjunction with HMMs for improving performance by training for improved performance. One example includes training on phrases that might be expected to be heard in voicemail systems such as “for faster service, please have your account number ready”, “press <digit> to return to the main menu”, etc. Another example is by training on the websites and/or (successfully recognized) voice menus of other businesses in the same category of the Yellow Pages as the business that is being called. Another example is by exploiting the “parallel speech corpora” that may be found in a menu that includes multiple language options such as English and Spanish. The two language subtrees may be expected to give the same options and limited machine translation may be used to help verify the correctness in understanding and/or fill in the gaps. In one embodiment, automatic speech recognition programs may be used along with manual corrections for some of these examples.
If this option does not go back to a previous menu (334), then it is determined whether this option has already been selected (340).
If this option has already been selected (340), then it is determined whether exploration of this option is complete (342). For example, this option may have been selected in a previous rendition and there may be another layer of choices offered in the voicemail system. If a second layer of choices associated with this option has not been fully explored, then the exploration of this option is not yet complete. If, however, every choice offered under this option has been explored, then the exploration of this option is complete.
If the exploration of this option is not yet complete (342), then this option is selected (352). If the exploration of this option is complete (342) then it is determined whether all options at this level have been explored (346). For example, if a voicemail system has two levels of choices such as the first level of choices being press 1 for English and 2 for Spanish, and then within the English menu having another level of choices such as press 1 for billing questions and 2 for new services, then all options at the first level has been explored if both the English menu and the Spanish have been explored. If all options at this level have been explored (346), then the call is terminated (348). If, however, all options at this level have not yet been explored (346), then a new option is selected to be analyzed (332 of
If this option has not already been selected (340), then it is determined whether this option will send the call to a person (350). One example of determining whether selecting this option will send a call to a person is to determine if the message associated with this option includes certain phrases. For example, if the option states “for X, press 1” and X is recognized as a name, then selecting the option may send the call to a person. One way of recognizing whether X is a person it to refer to a database of names such that names like Susan and John maybe recognized. Another way of recognizing whether X is a name is to correlate X with names that maybe listed on a website associated with this business. For example, if the phone number is for a medical center, the doctors are often listed on the medical center's website. If X matches a name listed on the website, then it can be determined that selecting this option will send the call to that person. Yet another example of recognizing whether X is a person is to detect common titles, such as “Dr.” or “Ms.” at the beginning of the name, or common morphemes in the surname, such as an initial “Mc” or “van”, or a final “son” or “owitz”, or common surnames overall, such as “Johnson”, “Smith”, or “Brown”. Additionally, there may be various signals that may indicate that a human is about to answer the call once an option has been selected. Examples include Muzak or a similarly recognizable audible entertainment for those on hold, or a message such as “Your call will be answered in the order in which it was received”, or “Your call will be monitored for your protection”.
If selecting the option sends a call to a person (350), then a new option is selected to be analyzed (332 of
The procedures for a person answering a call within a voicemail system maybe the same or different from the procedures from a person answering a call outside of a voicemail system. For example, if a new number is called and a person answers that call, the procedure maybe to ask the person to confirm the name of the business, whereas if a voicemail system answers the call and after choosing at least one option a person answers the call, the procedure maybe to terminate the call.
If a person does not answer the call (352), then the message is recorded (354). This phone number is redialed (356) and a option is selected for analysis (332).
The examples of
In one embodiment, a particular voicemail option, or a sequence of digits to choose that option, may be displayed rather than the entire voicemail tree. For example, a user may search for the terms “Mega Phone, speak to person”, in which case a sequence of numbers such as (123) 543-6975-1-0 may be displayed which includes Mega Phone's phone number as well as sequential selections of options 1 and 0. As shown in the example of
Although the foregoing embodiments have been described in some detail for purposes of clarity of understanding, the invention is not limited to the details provided. There are many alternative ways of implementing the invention. The disclosed embodiments are illustrative and not restrictive.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4674116 *||Jul 3, 1985||Jun 16, 1987||Amtelco||Telephone line concentrator without switching matrix|
|US5524137 *||Oct 4, 1993||Jun 4, 1996||At&T Corp.||Multi-media messaging system|
|US5799077 *||Feb 13, 1995||Aug 25, 1998||Fujitsu Limited||Method of and apparatus for automatic dialing|
|US6249809 *||Mar 4, 1998||Jun 19, 2001||William L. Bro||Automated and interactive telecommunications system|
|US6285362 *||Oct 1, 1999||Sep 4, 2001||Fujitsu Limited||Communication terminal and its display control system|
|US6542584 *||Aug 31, 1998||Apr 1, 2003||Intel Corporation||Digital telephone system with automatic voice mail redirection|
|US6633397 *||Jul 28, 1995||Oct 14, 2003||Canon Kabushiki Kaisha||Output apparatus and output environment setting method in output apparatus|
|US6909780 *||Jun 8, 2000||Jun 21, 2005||Altigen Communications, Inc.||Voice mail call out method and apparatus|
|US7065188 *||Oct 19, 1999||Jun 20, 2006||International Business Machines Corporation||System and method for personalizing dialogue menu for an interactive voice response system|
|US7123699 *||Aug 2, 2002||Oct 17, 2006||Estech Systems, Inc.||Voice mail in a voice over IP telephone system|
|US7145998 *||Jul 29, 2002||Dec 5, 2006||Bellsouth Intellectual Property Corporation||Systems, methods, and devices for a callback/callback reminder feature|
|US7167832 *||Oct 11, 2002||Jan 23, 2007||At&T Corp.||Method for dialog management|
|US7260537 *||Mar 25, 2003||Aug 21, 2007||International Business Machines Corporation||Disambiguating results within a speech based IVR session|
|US20030074091 *||Oct 11, 2002||Apr 17, 2003||Lg Electronics Inc.||User interface method and apparatus for appliance connected with host system|
|US20030171926 *||Mar 27, 2002||Sep 11, 2003||Narasimha Suresh||System for information storage, retrieval and voice based content search and methods thereof|
|US20040193403 *||Mar 25, 2003||Sep 30, 2004||International Business Machines Corporation||Disambiguating results within a speech based IVR session|
|US20050081152 *||Sep 25, 2003||Apr 14, 2005||International Business Machines Corporation||Help option enhancement for interactive voice response systems|
|US20050100145 *||Sep 14, 2004||May 12, 2005||Spencer Bradford L.||Multi-user intelligent call screening|
|US20050123118 *||Sep 29, 2004||Jun 9, 2005||Terry George A.||Dynamic call response system|
|US20060165104 *||Nov 10, 2004||Jul 27, 2006||Kaye Elazar M||Content management interface|
|1||Scansoft, Call Navigator, Reduce Costs & Maximize Customer Satisfaction through Natural Language, 2005, http://www.scansoft.com/speechworks/callnavigator/.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8290125 *||Sep 2, 2008||Oct 16, 2012||International Business Machines Corporation||Voice response unit shortcutting|
|US8340969 *||Apr 24, 2009||Dec 25, 2012||Research In Motion Limited||Method and mobile communication device for generating dual-tone multi-frequency (DTMF) commands on a mobile communication device having a touchscreen|
|US8515028 *||Mar 10, 2009||Aug 20, 2013||Foncloud, Inc.||System and method for externally mapping an Interactive Voice Response menu|
|US8605868 *||Jul 17, 2013||Dec 10, 2013||Foncloud, Inc.||System and method for externally mapping an interactive voice response menu|
|US8615396 *||Sep 2, 2008||Dec 24, 2013||International Business Machines Corporation||Voice response unit mapping|
|US8634521||Aug 21, 2012||Jan 21, 2014||International Business Machines Corporation||Voice response unit shortcutting|
|US8774373 *||Jul 16, 2013||Jul 8, 2014||Foncloud, Inc.||System and method for externally mapping an interactive voice response menu|
|US8908847||Mar 10, 2009||Dec 9, 2014||Foncloud, Inc.||System and method for deep dialing phone systems|
|US9003300||Oct 3, 2008||Apr 7, 2015||International Business Machines Corporation||Voice response unit proxy utilizing dynamic web interaction|
|US9014351||Nov 3, 2014||Apr 21, 2015||Foncloud, Inc.||System and method for deep dialing phone systems|
|US9100479 *||Mar 12, 2012||Aug 4, 2015||Angel.Com Incorporated||Answering machine detection|
|US9106745||Sep 16, 2008||Aug 11, 2015||International Business Machines Corporation||Voice response unit harvesting|
|US9270817||Nov 24, 2008||Feb 23, 2016||Foncloud, Inc.||Method for determining the on-hold status in a call|
|US9288316||Mar 3, 2009||Mar 15, 2016||Foncloud, Inc.||System and method for eliminating hold-time in phone calls|
|US9386151||Jun 9, 2014||Jul 5, 2016||Foncloud, Inc.||System and method for replacing hold-time with a call-back in a contact center environment|
|US9578170||Aug 3, 2015||Feb 21, 2017||Genesys Telecommunications Laboratories, Inc.||Answering machine detection|
|US20090136014 *||Nov 24, 2008||May 28, 2009||Foncloud, Inc.||Method for Determining the On-Hold Status in a Call|
|US20090202050 *||Mar 10, 2009||Aug 13, 2009||Foncloud, Inc.||System and method for deep dialing phone systems|
|US20090207980 *||Mar 10, 2009||Aug 20, 2009||Foncloud, Inc.||System and method for externally mapping an interactive voice response menu|
|US20090207996 *||Mar 3, 2009||Aug 20, 2009||Foncloud, Inc.||System and method for eliminating hold-time in phone calls|
|US20100054430 *||Sep 2, 2008||Mar 4, 2010||Grigsby Travis M||Voice response unit shortcutting|
|US20100057456 *||Sep 2, 2008||Mar 4, 2010||Grigsby Travis M||Voice response unit mapping|
|US20100067670 *||Sep 16, 2008||Mar 18, 2010||Grigsby Travis M||Voice response unit harvesting|
|US20100088613 *||Oct 3, 2008||Apr 8, 2010||Lisa Seacat Deluca||Voice response unit proxy utilizing dynamic web interaction|
|US20100274563 *||Apr 24, 2009||Oct 28, 2010||Research In Motion Limited||Method and mobile communication device for generating dual-tone multi-frequency (dtmf) commands on a mobile communication device having a touchscreen|
|US20120230483 *||Mar 12, 2012||Sep 13, 2012||Angel.Com||Answering machine detection|
|US20130301814 *||Jul 16, 2013||Nov 14, 2013||Foncloud, Inc.||System and method for externally mapping an interactive voice response menu|
|U.S. Classification||379/88.18, 370/401, 379/211.02, 715/733, 709/217, 704/201, 379/88.23, 700/83, 379/266.07, 715/705, 704/270.1|
|Cooperative Classification||H04M2203/2027, H04M3/533|
|Sep 6, 2005||AS||Assignment|
Owner name: GOOGLE, INC.,CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:GOLDING, ANDREW R.;PASZTOR, EGON;REEL/FRAME:016495/0830
Effective date: 20050808
|Nov 12, 2013||FPAY||Fee payment|
Year of fee payment: 4